{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XJ3ECPGIWZES5SLKFN4V6NXCLB","short_pith_number":"pith:XJ3ECPGI","schema_version":"1.0","canonical_sha256":"ba76413cc8b6492ec96a2b795f36e2586e37d9265b052f8cb1ec8665d28d9238","source":{"kind":"arxiv","id":"2103.08849","version":3},"attestation_state":"computed","paper":{"title":"Multilingual Multimodal Pre-training for Zero-Shot Cross-Lingual Transfer of Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Alexander Hauptmann, Florian Metze, Graham Neubig, Junjie Hu, Mandela Patrick, Po-Yao Huang","submitted_at":"2021-03-16T04:37:40Z","abstract_excerpt":"This paper studies zero-shot cross-lingual transfer of vision-language models. Specifically, we focus on multilingual text-to-video search and propose a Transformer-based model that learns contextualized multilingual multimodal embeddings. Under a zero-shot setting, we empirically demonstrate that performance degrades significantly when we query the multilingual text-video model with non-English sentences. To address this problem, we introduce a multilingual multimodal pre-training strategy, and collect a new multilingual instructional video dataset (MultiHowTo100M) for pre-training. Experimen"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2103.08849","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-03-16T04:37:40Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"525d9274bb94cd1775d39419b9688d25dde3f6b8619ea600aa75c72cdf4d1097","abstract_canon_sha256":"9f75df1588d7c2253839d2d06a1a5ad20c6efc6db7c3ebb5f7e034cfc5e794c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:32:16.679188Z","signature_b64":"TPgWXPGOzD0TWd7e+HipovgO2lNT3QvZGQJGO3ONUfQBVi4r9RVc6ePf6SagedvfE/UyftRaFwRiPabxZ7vbBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba76413cc8b6492ec96a2b795f36e2586e37d9265b052f8cb1ec8665d28d9238","last_reissued_at":"2026-07-05T02:32:16.678772Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:32:16.678772Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Multimodal Pre-training for Zero-Shot Cross-Lingual Transfer of Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Alexander Hauptmann, Florian Metze, Graham Neubig, Junjie Hu, Mandela Patrick, Po-Yao Huang","submitted_at":"2021-03-16T04:37:40Z","abstract_excerpt":"This paper studies zero-shot cross-lingual transfer of vision-language models. Specifically, we focus on multilingual text-to-video search and propose a Transformer-based model that learns contextualized multilingual multimodal embeddings. Under a zero-shot setting, we empirically demonstrate that performance degrades significantly when we query the multilingual text-video model with non-English sentences. To address this problem, we introduce a multilingual multimodal pre-training strategy, and collect a new multilingual instructional video dataset (MultiHowTo100M) for pre-training. Experimen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.08849","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2103.08849/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2103.08849","created_at":"2026-07-05T02:32:16.678830+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.08849v3","created_at":"2026-07-05T02:32:16.678830+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.08849","created_at":"2026-07-05T02:32:16.678830+00:00"},{"alias_kind":"pith_short_12","alias_value":"XJ3ECPGIWZES","created_at":"2026-07-05T02:32:16.678830+00:00"},{"alias_kind":"pith_short_16","alias_value":"XJ3ECPGIWZES5SLK","created_at":"2026-07-05T02:32:16.678830+00:00"},{"alias_kind":"pith_short_8","alias_value":"XJ3ECPGI","created_at":"2026-07-05T02:32:16.678830+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB","json":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB.json","graph_json":"https://pith.science/api/pith-number/XJ3ECPGIWZES5SLKFN4V6NXCLB/graph.json","events_json":"https://pith.science/api/pith-number/XJ3ECPGIWZES5SLKFN4V6NXCLB/events.json","paper":"https://pith.science/paper/XJ3ECPGI"},"agent_actions":{"view_html":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB","download_json":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB.json","view_paper":"https://pith.science/paper/XJ3ECPGI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.08849&json=true","fetch_graph":"https://pith.science/api/pith-number/XJ3ECPGIWZES5SLKFN4V6NXCLB/graph.json","fetch_events":"https://pith.science/api/pith-number/XJ3ECPGIWZES5SLKFN4V6NXCLB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB/action/storage_attestation","attest_author":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB/action/author_attestation","sign_citation":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB/action/citation_signature","submit_replication":"https://pith.science/pith/XJ3ECPGIWZES5SLKFN4V6NXCLB/action/replication_record"}},"created_at":"2026-07-05T02:32:16.678830+00:00","updated_at":"2026-07-05T02:32:16.678830+00:00"}